The Research Of The Parameter's Optimization Of The Adaptive Controller

In all kinds of control system, because the control object has some uncertainty and various disturbance. Including many complex factors, such as ,environment change ,parameter change and so on. which cause the control effect cann't meet the request of system performance. The adaptive control auto regulate the controller parameters to eliminate the complex factors and uncertainty influence, In order to make the controller and control object to adapt to the environment, but it just suitable for the parameters change little. In essence, adaptive control overcomes and compensates uncertainty and disturbance by estimating the factors relate to system performance.Intelligent control analyzes and synthesizes system from system performance and global optimization to achieves the best control effects. It can be divided into fuzzy control, neural network control and so on. It can be used as compensation link of the system with complexity and uncertainty. It also can identify and control the nonlinear system. And even can optimizes the control. So the intelligent control has wide application in adaptive control. This paper proposes a method of adaptive control combines with intelligent control.The neural network has self-learning function. It can approach to any nonlinear function. So, It can be used as compensation link of the system with complexity and uncertainty. It also can identify and control the nonlinear system. And even can optimizes the control. So, the adaptive control based on neural network has important research meanings.The fuzzy control is a efficient method to solve the inference system or the system with imprecise and uncertainty. When fuzzy control is used in the complex system with time-varied and nonlinear uncertainty, in order to gain good control effects, It is asked to have complete control rules. These control rules are induction of fuzzy information and summarization of operation experience. Because of nonlinear and high-order characteristic and time-varied and random of the control process, leading to the control rules are rough and insufficient more or less. So as to compensate this disadvantage, fuzzy control develop toward the direction of adaptive and auto-organization and self-learning. Which make fuzzy control rules self-adjust and self-modify and self-optimize in control process. So that the system's performance improve continuously, achieve the best effects. So, the adaptive control based on fuzzy logical has important research meanings.Finally, we make simulation in several real plants by using the methods proposed above. The result is good. Show the scheme of adaptive control combines with intelligent control is feasible and ideal.